PETRONAS ROTATING EQUIPMENT ANALYTICS (PROTEAN) Presented by Khairil Azwan B Khabri COE Upstream Gavin Halls COE Upstream 2017 PETROLIAM NASIONAL BERHAD (PETRONAS) All rights reserved. No part of this document may be reproduced, stored in a retrieval system or transmitted in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the permission of the copyright owner. 2017 Petroliam Nasional Berhad (PETRONAS) 1
Company Overview Petroleum Nasional Berhad (PETRONAS) is Malaysia s fully integrated oil & gas multinational with proven capabilities in a broad spectrum of the petroleum chain value. Establish in August 17, 1974, PETRONAS vision is to be a Leading Oil and Gas Multinational of Choice PETRONAS mission statements are We are a business entity Petroleum is our core business Our primary responsibility is to develop and add value to this national resource Our objective is to contribute to the well-being of the people and the nation 2017 Petroliam Nasional Berhad (PETRONAS) 2
PETRONAS Core Business 2017 Petroliam Nasional Berhad (PETRONAS) 3
Initial Problem Statement Over 130 pieces of gas turbine driven equipment Numerous manufacturers Centrifugal Gas Compressor Generator 30+ Pieces of Super Critical Equipment single duty high value Numerous Reciprocating engines and pumps OEM and 3 rd Party solutions mean financial commitment would be considerable Management directive to go digital 2017 Petroliam Nasional Berhad (PETRONAS)
When did all begin? PETRONAS Maintenance & Engineering team looked at how to perform diagnostics using unit data on one single unit cost effectively Team compiled monthly report using running data and equipment vibration and lube oil health status August 2015 December 2015 May 2015 Utilized Operation Data Management system (ODMS) which is a data capture system using handheld data loggers October 2015 Team developed the process flow and investigated ways to improve the overall monitoring system 2017 Petroliam Nasional Berhad (PETRONAS) 5
Understand data Monitoring actual data Monthly report was produced showing health of the unit Visualized data Captured a few issues Unable to run macro to bring data in from spreadsheet automatically Poor data entry by operators Numerous errors in data capture Manual process to capture data from ODMS Understand process flow 2017 Petroliam Nasional Berhad (PETRONAS) 6
Basic Design Brief Not to duplicate machinery HMI Not to duplicate the unit control system Powerful automatic analytical tools to identify issues before they become issues Simple visuals whilst providing as much relevant information to the engineer as possible Trends and graphs take human process power to look for issue Automated alert system Email Ability to review historical data Not limited to Major Rotating Equipment Support other disciplines Electrical health Pumps Vessels Ability to evolve system Visuals Algorithms Integrate with other systems ADD Value to PETRONAS 2017 Petroliam Nasional Berhad (PETRONAS) 7
2017 Petroliam Nasional Berhad (PETRONAS)
Lessons Learnt and Limitations Understanding the need for management of a system Very manual to understand alerts Excel not ideal as very processor heavy: constant No automatic email notifications Difficult to integrate into other PETRONAS systems 2017 Petroliam Nasional Berhad (PETRONAS) 9
System Solution Architecture 6 Informed Decisions Monitoring & Surveillance Analysis & Predictive Action & Optimization 5 Smart Visualization 4 Workflows & Integration Business Process Automation Engine Analysis & Predictive Engines Optimization Engines 3 Applications PI System 2 Data Management Historians 1 Data Sources Real Time 2017 Petroliam Nasional Berhad (PETRONAS) 10
Design Brief for next step November 2016 Clear aim on the system to be developed with element of People, process & Technology Visuals should be meaningful System does the hard work and complex thinking processes Link with other PETRONAS software solutions Human to make final decision 2017 Petroliam Nasional Berhad (PETRONAS) 11
How do OSIsoft help PROTEAN PI Data Archive Excel not ideal as very processor heavy: constant Very manual to understand alerts PI AF No automatic email notifications Notifications Asset Analytics No algorithm embedded to the systems PI Coresight Dashboard Difficult to integrate PI System into other Access PETRONAS systems 2017 Petroliam Nasional Berhad (PETRONAS) 12
Early development of PROTEAN 01 Developed system for 2 critical 02 gas turbine driven compressor units Utilized PETRONAS PI System for data collection and algorithms. Utilized PI Coresight for visuals Automatic e-mail notifications of issues proved difficult and unreliable. Email notification of unit shutdown successful 03 04 Data dump when unit shuts down to aid RCFA : CSV file on shared drive 05 Data points analyzed based on 06 criticality of data point Analytic page simple but effective 2017 Petroliam Nasional Berhad (PETRONAS) 13
Developed PROTEAN in PI System Environment 2017 Petroliam Nasional Berhad (PETRONAS) 14
Did it Work? Yes System alerted to numerous concerns over a 2 month period Email notifications worked Visuals worked well PROTEAN Dashboard 2017 Petroliam Nasional Berhad (PETRONAS) 15
During development and implementation May 2017 Increase number of units covered and ISO 14224 system classification Multiple data point comparisons Increase number of units covered Understanding of PI AF structure and power Better email notifications. More information to recipient. Utilized Event Frame process Create templates for algorithm. Reduce processor and server loading. Not all algorithms are required on all data points 2017 Petroliam Nasional Berhad (PETRONAS) 16
Current Look and Feel 2017 Petroliam Nasional Berhad (PETRONAS) 17
PROTEAN as tools in PETRONAS Integrated Operation (IO) Automatic Process Human to Human communication Reporting Automatic process E-mail alert to relevant parties Platform Central ICBM Department Regional Platform Automatic Process ICBM Manager Monthly Report Concerns ICBM Engineer Operations Unit Automatic data from PI E-mail alert to relevant parties Concerns Algorithms run ICBM Engineer SME Notification received Analysis output Initial actions carried out Unit Automatic data from PI Algorithms run Validation of data Analysis output Concerns Review data and confirm issue Validation of data Issue initial action or recommendation to respective stakeholders Critical Trends shown on IO Recommendations will show Issue Criticality Possible solutions Plan for rectification EIC If rectification is required Critical Trends shown on IO Raise SAP notification Maintenance Improvement in algorithms Closing the circle system improvements Confirmed by data Rectification carried out Closing the circle system improvements 2017 Petroliam Nasional Berhad (PETRONAS) 18
PROTEAN in demand! No No of of unit of unit in in December in October April 2017 2017 2016 2017 Petroliam Nasional Berhad (PETRONAS) 19
Next Development CMMS, PETRONAS Asset Management, online drawings and other system integrations Further Supercritical and C1 units to be incorporated 100 + units by 2019 Next Development Research more complex algorithms and OSIsoft PI System add-on Ability for individuals to run reports on equipment and compare between regions Technical Condition Index (TCI) for systems and unit for easy visualization of equipment health Develop fault tree to assist engineers with investigation and diagnosis of alerts Develop risk based maintenance philosophy (condition based) and move away from fixed periodic maintenance schedules 2017 Petroliam Nasional Berhad (PETRONAS) 20
Summary as of Q3 2017 PROTEAN is developing very well and meeting expectations Low cost and high return on investment predicted Evolution rather than revolution 270 man hours each unit to replicate due to template system PROTEAN future upgrades will be based on benefit to user 2017 Petroliam Nasional Berhad (PETRONAS) 21
The Future Pattern recognition Future data - prediction Analysis tools Baseline scoring Spare part optimization 2017 Petroliam Nasional Berhad (PETRONAS) 22
Final Thoughts Why pay the OEM for Remote Monitoring and Diagnostics (RM&D) when the OEM is receiving all the data for R&D? The OEM should be paying the operator for the data business model change? 2017 Petroliam Nasional Berhad (PETRONAS) 23
THE FUN PART IS WE DESIGNED A FANTASTIC LOGO!!!! Symbolizes the rotating element Symbolizes the machinery heartbeat Trident for Predictive Diagnostic Prognostic 2017 Petroliam Nasional Berhad (PETRONAS) 24
Khairil Azwan Bin Khabri khairilazwan@petronas.com Head Reliability Management Operational Excellence Center of Excellence PETRONAS Carigali Gavin Warren Halls Gavinwarren.halls@petronas.com Rotating Machinery Specialist Operational Excellence Center of Excellence PETRONAS Carigali Nazrul Hsyraf M Mokhtar nazrul.mokhtar@petronas.com.my PI Specialist Group Technical Data Project Delivery and Technology PETRONAS 25 2017 Petroliam Nasional Berhad (PETRONAS) 25
Questions Please wait for the microphone before asking your questions Please remember to Complete the Online Survey for this session State your name & company EMEA USERS CONFERENCE 2017 LONDON #OSISOFTUC 2017 OSIsoft, LLC 26
Siti Zainah Mapak Abayazi B Zainul Gavin Warren Halls M Hafiz B Husin Nur Sharafina Md Abidin Shah Nabilah Nur Amanina Abdul Rahim Nazrul Hsyraf B M Mokhtar Anuar Ismail Madzalan B Nawi Saharudin B M Rejab Chen Kah Seong Muqrim Dzubairy B Shahaimi Sultan Maideen Noorul Amen Khairul Rijal Erma Fathin Aqera Nor Affendi B Jemidi M Faizal B Mohamed Josiah Tay Guang Wei Nisa Balqis Masnoor Zaidah Zainal Sharizan Ramli Muhammad Hafizi Mat Saat M Muslim B Jusoh M Zaha Mira B M Idrus Azlan B Ayub M Nazori B Janor Shaharuddin Hamid Mustapha Musreen Azwan Mustafa A Halim B A Gani Khairul Anwar B M Nor M Fadzly B M Khalil M Shafek A Aziz Idris Ibrahim Rahmat Nur Adi Wijaya Hairil Helmy Hamzah Marina Syazlin M Saidi Hew Pei Chin A Ghaffar Dawam Arif Ibrahim M Farid B M Ismail M Shahrizal B Jasmani Megat Nazri B Hanifah Shairine Manlosa Syazwani Mohd Ali 2017 Petroliam Nasional Berhad (PETRONAS) 27
Thank You 2017 Petroliam Nasional Berhad (PETRONAS) 28